This invention relates to cancer detection and treatment and, more particularly, to a novel protein called xe2x80x9cp19ARF protein,xe2x80x9d xe2x80x9cp19ARFxe2x80x9d or simply xe2x80x9cARF-p19xe2x80x9d that is involved in regulation of the eukaryotic cell cycle. Protein ARF-p19 is encoded by a nucleic acid derived from the gene, INK4A, which also encodes an inhibitor of D-type cyclin-dependent kinases called xe2x80x9cp16Ink4a protein,xe2x80x9d xe2x80x9cp16InK4axe2x80x9d or simply xe2x80x9cInK4a-p16.xe2x80x9d
Transcripts encoding InK4a-p16 originate from a first promoter, E1xcex1; the present invention is based on the observation that some INK4A transcripts initiate from a second promoter, E1xcex2, and contain an Alternative Reading Frame, ARF, which overlaps the InK4a-p16 reading frame to some degree. ARF transcripts direct the production of a protein that has ARF-p19 amino acid sequences instead of the previously-known InK4a-p16 sequences. Like InK4a-p16, ARF-p19 regulates the eukaryotic cell cycle. When overexpressed, ARF-p19 inhibits cells from proceeding past both the G1 and G2 phases of the cell cycle. However, the mechanism(s) by which ARF-p19 acts are unlike those of InK4a-p16, which acts by directly and specifically interacting with CDK (cyclin D-dependent kinase) proteins and thus preventing CDK-cyclin D interactions.
In addition to (1) ARF-p19 proteins, this invention further relates to (2) nucleic acids that encode ARF-p19 isolated from mice, humans and other mammals; (3) antibodies that specifically bind ARF-p19 protein or polypeptides derived therefrom; (4) methods for detecting one or more nucleic acids encoding ARF-p19, or alterations in such nucleic acids; (5) methods for producing ARF-p19 proteins using nucleic acids that encode ARF-p19; (6) purified ARF-p19 proteins, or fusion proteins derived from the joining of an ARF-p19 polypeptide sequence with a second polypeptide sequence; (7) methods of treating cancer using purified ARF-p19 proteins or fusion proteins derived therefrom; (8) methods of inducing cell cycle arrest using ARF-pi9 proteins or nucleic acids encoding ARF-p19 proteins; (9) methods for detecting ARF-p19 proteins using antibodies that specifically bind ARF-p19 proteins; (10) methods of selectively killing cells having uncontrolled growth using antibodies that specifically bind ARF-p19 proteins, or conjugates derived from such antibodies; (11) methods of stimulating cell growth using antibodies that specifically bind ARF-p19 proteins, or fragments derived from such antibodies; and (12) transgenic non-human animals that have a genetically engineered alteration in one or more nucleic acids encoding ARF-p19 proteins but which express normal levels of wild-type InK4a-p16 protein, or which overexpress human ARF-p19 or mutant forms of ARF-p19.
Neoplasia, the pathological process by which tumors develop, necessarily involves unregulated, or at best misregulated, cellular growth and division. The molecular pathways that regulate cellular growth must inevitably intersect with those that regulate the cell cycle. The cell cycle consists of a cell division phase and the events that occur during the period between successive cell divisions, known as interphase. Interphase is composed of successive G1, S, and G2 phases, and normally comprises 90% or more of the total cell cycle time. Most cell components are made continuously throughout interphase; it is therefore difficult to define distinct stages in the progression of the growing cell through interphase. One exception is DNA synthesis, since the DNA in the cell nucleus is replicated only during a limited portion of interphase. This period is denoted as the S phase (S=synthesis) of the cell cycle. The other distinct stage of the cell cycle is the cell division phase, which includes both nuclear division (mitosis) and the cytoplasmic division (cytokinesis) that follows. The entire cell division phase is denoted as the M phase (M=mitotic). This leaves the period between the M phase and the start of DNA synthesis, which is called the G1 phase (G=gap), and the period between the completion of DNA synthesis and the next M phase, which is called the G2 phase (Alberts, B. et al., Molecular Biology of the Cell, Garland Publishing, Inc., New York and London (1983), pages 611-612.).
Progression through different transitions in the eukaryotic cell cycle is positively regulated by a family of master enzymes, the cyclin-dependent kinases (reviewed by Sherr, C. J., Cell 73:1059-1065 (1993)). These holoenzymes are composed of two proteins, a regulatory subunit (the cyclin), and an associated catalytic subunit (the actual cyclin-dependent kinase or CDK), the levels of which vary with different phases of the cell cycle (Peters, G., Nature 371:204-205 (1994)). Both cyclins and CDKs represent molecular families that encompass a variety of genetically related but functionally distinct proteins. Generally, different types of cyclins are designated by letters (i.e., cyclin A, cyclin B, cyclin D, cyclin E, etc.); CDKs are distinguished by numbers (CDK1, CDK2, CDK3, CDK4, CDK5, etc.; CDK1 is a.k.a. CDC2).
CDK-cyclin D complexes regulate the decision of cells to replicate their chromosomal DNA (Sherr, Cell 73:1059-1065 (1993)). As cells enter the cycle from quiescence, the accumulation of CDK-cyclin D holoenzymes occurs in response to mitogenic stimulation, with their kinase activities being first detected in mid-G1 phase and increasing as cells approach the G1/S boundary (Matsushime et al., Mol. Cell. Biol. 14:2066-2076 (1994); Meyerson and Harlow, Mol. Cell. Biol. 14:2077-2086 (1994)). The cyclin D regulatory subunits are highly labile, and premature withdrawal of growth factors in G1 phase results in a rapid decay of CDK-cyclin D activity that correlates with the failure to enter S phase. In contrast, removal of growth factors late in G1 phase, although resulting in a similar collapse of CDK-cyclin D activity, has no effect on further progression through the cell cycle (Matsushime et al., Cell 65:701-713 (1991)). Microinjection of antibodies to cyclin D1 into fibroblasts during G1 prevents entry into the S phase, but injections performed at or after the G1xe2x86x92S transition are without effect (Baldin et al., Genes and Devel. 7:812-821 (1993); Quelle et al., Genes and Devel. 7:1559-1571 (1993)). Therefore, CDK-cyclin D complexes execute their critical functions at a late G1 checkpoint, after which cells become independent of mitogens for completion of the cycle.
In mammals, cells enter the cell cycle and progress through G1 phase in response to extracellular growth signals which trigger the transcriptional induction of D-type cyclins. The accumulation of D cyclins leads to their association with two distinct catalytic partners, CDK4 and CDK6, to form kinase holoenzymes. Several observations argue for a significant role of the cyclin D-dependent kinases in phosphorylating the retinoblastoma protein, pRb, leading to the release of pRB-associated transcription factors that are necessary to facilitate progression through the G1xe2x86x92S transition. First, CDK-cyclin D complexes have a distinct substrate preference for pRb but do not phosphorylate the canonical CDK substrate, histone H1 (Matsushime et al., Cell 71:323-334 (1992); Matsushime et al., Mol. Cell. Biol. 14:2066-2076 (1994); Meyerson and Harlow, Mol. Cell. Biol. 14:2077-2086 (1994)). Their substrate specificity may be mediated in part by the ability of D-type cyclins to bind to pRb directly, an interaction which is facilitated by a Leu-X-Cys-X-Glu pentapeptide that the D cyclins share with DNA oncoproteins that also bind pRb (Dowdy et al., Cell 73:499-511 (1993); Ewen et al., Cell 73:487-497 (1993); Kato et al., Genes and Devel. 7:331-342 (1993)). Second, cells in which pRb function has been disrupted by mutation, deletion, or after transformation by DNA tumor viruses are no longer inhibited from entering S phase by microinjection of antibodies to D cyclin, indicating that they have lost their dependency on the cyclin D-regulated G1 checkpoint (Lukas et al., J. Cell. Biol. 125:625-638 (1994); Tam et al., Oncogene 9:2663-2674 (1994)). However, introduction of pRb into such cells restores their requirement for cyclin D function (Lukas et al., J. Cell. Biol. 125:625-638 (1994)). Third, pRb-negative cells synthesize elevated levels of a 16 kDa polypeptide inhibitor of CDK4, xe2x80x9cp16InK4axe2x80x9d (a.k.a. xe2x80x9cInK4a-p16xe2x80x9d or simply xe2x80x9cp16xe2x80x9d), which is a member of a recently discovered class of cell cycle regulatory proteins (Nasmyth and Hunt, Nature 366:634-635 (1993); Peters, G., Nature 371:204-205 (1994)) and which is found in complexes with CDK4 at the expense of D-type cyclins during G1 phase (Bates et al., Oncogene 9:1633-1640 (1994); Serrano et al., Nature 366:704-707 (1993); Xiong et al., Genes and Devel. 7:1572-1583 (1993)). The fact that such cells cycle in the face of apparent CDK4 inhibition again implies that D-type cyclins are dispensable in the Rb-negative setting.
The InK4 gene family (xe2x80x9cInK4xe2x80x9d signifies Inhibitors of CDK4) is known to include at least three other low molecular weight polypeptides, InK4b-p15, induced in human epithelial cells treated by transforming growth factory-xcex2 (TGF-xcex2) (Hannon, G. J., and Beach, D., Nature 371:257-261 (1994)), InK4d-p19 (Hirai, H., et al., Mol. Cell. Biol. 15:2672-2681 (1995)) and InK4c-p18 (Guan et al., Genes and Develop. 8:2939-2952 (1994); Hirai, H., et al., Mol. Cell. Biol. 15:2672-2681 (1995)). InK4d-p19 and InK4c-p18 are described in detail in Ser. No. 08/384,106, filed Feb. 6, 1995, which is hereby incorporated by reference.
Members of the InK4 family are typically composed of repeated ankyrin motifs, each of about 32 amino acids in length. All known members of the InK4 family act to specifically inhibit enzymatic activities of D-type cyclin-dependent kinases such as CDK4 and CDK6. Unlike other universal CDK inhibitors, such as p21Cip1/Waf1 (E1-Deiry et al., Cell 75:817-825 (1993); Gu et al., Nature 366:707-710 (1993); Harper et al., Cell 75:805-816 (1993); Xiong et al., Nature 366:701-704 (1993)) and p27Kip1 (Polyak et al., Genes and Devel. 8:9-22 (1994); Polyak et al., Cell 78:59-66 (1994); Toyoshima and Hunter, Cell 78:67-74 (1994)), the InK4 proteins selectively inhibit the activities of CDK4 and CDK6, but do not inhibit the activities of other CDKs (Guan et al., Genes and Devel. 8:2939-2952 (1994); Hannon and Beach, Nature 371:257-261 (1994); Serrano et al., Nature 366:704-707 (1993)).
Like many CDK inhibitors (CKIs) (Nasmyth and Hunt, Nature 366:634-635 (1993)), InK4 family members negatively regulate progression through the mammalian cell cycle, in part in response to anti-proliferative extracellular signals. The InK4 proteins, by inhibiting the activities of a specific class of the D-type cyclin-dependent kinases (i.e., CDK4 and/or CDK6), arrest cell cycle progression in G1 phase and thus prevent cells from replicating their chromosomal DNA. Thus, in contradistinction to the positive regulation of D-type cyclin synthesis by growth factors, extracellular inhibitors of G1 progression can negatively regulate the activity of D-type cyclin-dependent kinases by inducing InK4 proteins.
Mullis et al., U.S. Pat. No. 4,965,188 (Oct. 23, 1990), describe methods for amplifying nucleic acid sequences using the polymerase chain reaction (PCR).
Beach, published PCT patent application WO 92/20796 (Nov. 26, 1992), describes genes encoding D cyclins and uses thereof.
Berns, U.S. Pat. No. 5,174,986 (Dec. 29, 1992), describes methods for determining the oncogenic potential of chemical compounds using a transgenic mouse predisposed to develop T-cell lymphomas.
Crissman et al., U.S. Pat. No. 5,185,260 (Feb. 9, 1993), describe methods for distinguishing and selectively killing transformed (neoplastic) cells using synthetic G1 kinase inhibitors.
Stone, S., et al., Cancer Research 55:2988-2994 (1995), describe two cDNAs derived from the human INK4A gene, including an xe2x80x9cxcex1formxe2x80x9d encoding InK4a-p16 and xcex2xe2x80x9c, formxe2x80x9d that includes an open reading frame (designated xe2x80x9cORF 2xe2x80x9d) that overlaps the reading frame encoding the ARF-p19 protein described herein. Stone et al. state that it xe2x80x9cis unknown if ORF 2 encodes a proteinxe2x80x9d (legend to FIG. 1, page 2990) and indicate that xe2x80x9cORF 2 has not been selectively maintained and probably does not encode a proteinxe2x80x9d (page 2989, column 2, lines 20-21).
Mao, L., et al., Cancer Research 55:2995-2997 (1995), describe two transcripts and corresponding cDNAs derived from the human INK4A gene, designated xe2x80x9cp16xe2x80x9d and xe2x80x9cp16xcex2.xe2x80x9d The p16 transcript is stated to encode the InK4a-p16 protein, while the p16xcex2 transcript is stated to contain a xe2x80x9ctheoretical open reading framexe2x80x9d (page 1996, column 1, line 47) that is not further defined, and suggest this sequence xe2x80x9cprobably represents an untranslated open reading framexe2x80x9d (page 2997, column 2, lines 9-10). Mao et al. state that the in vitro transcription and translation (TNT) product of the p16xcex2 cDNA is recognized by an antibody to InK4a-p16 polypeptide sequences (page 2997, column 1, lines 6), suggesting that the p16xcex2 transcript encodes an amino-terminal truncated InK4a-p16 polypeptide rather than a protein having, as ARF-p19 does, an amino acid sequence unrelated to that of InK4a-p16. However, Mao et al. also state that, using InK4a-p16 antiserum, they are unable to identify an amino-terminal truncated p16xcex2 protein in cell lines (page 2997, column 2, lines 1-2). Thus, Mao et al. are silent regarding the ARF-p19 protein described herein.
The present invention relates to the discovery in mammalian cells of a novel cell cycle regulatory protein, having a predicted molecular mass of 19 kDa, here designated xe2x80x9cARF-p19 proteinxe2x80x9d or more simply xe2x80x9cARF-p19xe2x80x9d. In particular, the invention relates to ARF-p19 proteins isolated from cells derived from a mouse or a human. Although derived from the gene encoding the previously-known InK4a-p16 protein, ARF-p19 arises by differential transcription and translation of InK4a-p16 sequences. That is, ARF-p19 is encoded by an alternative reading frame (ARF) and has an amino acid sequence (SEQ ID NO:2; SEQ ID NO:4) that is wholly unrelated to that of InK4a-p16. Surprisingly, however, ARF-p19 protein functions to regulate the cell cycle in a similar but less specific manner than, and by a mechanism distinct from that of, InK4a-p16 protein.
Thus, one aspect of the invention is directed to methods of using the ARF-p19 proteins of the invention to inhibit the growth of cancer cells and/or to prevent cancer cells from replicating their chromosomal DNA. Both InK4-p16 and InK4-p15 appear to act as tumor suppressors (Noburi, T. et al., Nature 368:753-756 (1994); Karnb, A. et al., Science 264:436440 (1994)). The genes encoding p16 and p15 map in a tandem array to the short arm of human chromosome 9 within a region that is frequently deleted in cancer cells, and the resulting loss of their anti-proliferative functions can contribute to tumorigenesis (Noburi et al., Nature 368:753-756 (1994); Okuda, T., et al., Blood 85:2321-2330 (1995)). The novel ARF-p19 protein described herein (1) plays a role in preventing the G1xe2x86x92S and G2xe2x86x92M phase transitions in normal mammalian cells, and (2) if having reduced or altered activity due to one or more mutations affecting the alternative reading frame encoding ARF-p19, could contribute to oncogenesis in some cancers, even if such mutations have no effect on the reading frame encoding InK4a-p16.
In another aspect, the invention provides nucleic acid sequences encoding ARF-p19 polypeptides from mice, humans and other mammals. The nucleic acid sequences of the invention may be expressed in the form of isolated nucleic acids, such as cDNA clones, genomic DNA clones, MRNA transcribed from either cDNA or genomic DNA clones, synthetic oligonucleotides, and/or synthetic amplification products resulting from PCR, and may be single-stranded or double-stranded.
In a related aspect, the invention provides methods for detecting nucleic acids encoding wild-type or mutant ARF-p19 proteins using the nucleic acid sequences of the invention described above. The detection of point mutations, deletions of, or other mutations in, the reading frame encoding ARF-p19 is predictive of a predisposition to, or diagnostic of, certain types of cancer.
In another related aspect, the DNA molecules of the invention described above may be cloned into expression vectors and placed in an appropriate host in order to produce ARF-p19 proteins or fusion proteins containing ARF-p19 polypeptide sequences. When placed in an animal that has cancer, this aspect of the invention relates to gene therapy for certain types of cancers.
In another aspect, the invention provides antibody compositions that bind specifically to ARF-p19 proteins and/or polypeptides derived therefrom. The antibody compositions of the invention may be polyclonal, monoclonal, or monospecific. Although all of the antibody compositions of the invention bind specifically to ARF-p19, some compositions bind to a specific epitope of ARF-p19 and thereby inhibit a specific function of ARF-p19.
In a related aspect, the invention provides methods for detecting ARF-p19 proteins using the antibody compositions described above. The detection of reduced amounts of, or altered forms of, ARF-p19 proteins is predictive of a predisposition to, or diagnostic of, certain types of cancer.
In another aspect, the invention provides transgenic non-human animals which have one or more mutations in the endogenous reading frame encoding ARF-p19, wherein said mutation results in the production of a mutant ARF-p19 protein or results in a loss of ARF-p19 expression but does not significantly affect the InK4a-p16 gene product or expression thereof. Additionally or alternatively, the transgenic non-human animals of the invention express a human wild-type or mutant ARF-p19. Because of the transgene(s) introduced into the genome of the non-human animals of the invention, the animals have a reduced and/or altered ARF-p19 activity compared to wild-type animals, and consequentially develops certain types of cancers, particularly melanomas, in a reproducible and thus predictable manner.
In a related aspect, compositions are evaluated for their potential to enhance or inhibit certain types of cancers, particularly melanomas, using the transgenic non-human animals of the invention.